niallcaldwell

Thanks 3D and Phil for the replies so far.

I can now see the advantages of the programmable servos like Hitec, not just for tuning the position loop but also for range and offset adjustment.

So it seems that they have just about enough resolution, but that the position loop may not be stiff enough to achieve the full accuracy under load. The servo will be slaved to a microcontroller which can sample the throttle position sensor, so it's possible that I could do the integration as a seperate outer loop and modify the servo command signal to achieve zero steady state error. This would not be difficult as sample speed for integration is much slower than for proportional.
But hobby servos might be lacking in terms of reliability, especially in the harsh environment of the underbonnet.

I suppose I could drive the throttle via a cable from the much better environment of the cabin. But a spring would be needed to keep the cable taught, which would increase current consumption.

hmmmm...